Changeset 8218 for trunk/gsdl/perllib/plugins/PDFPlug.pm

Timestamp:

2004-10-05T12:39:06+13:00 (20 years ago)

Author:

jrm21

Message:

use the unicode::ensure_utf8() function on the extracted text so we can
be sure that we don't generate bad archive files.

(BasPlug already does that when adding metadata.)

File:

• trunk/gsdl/perllib/plugins/PDFPlug.pm (modified) (4 diffs)

trunk/gsdl/perllib/plugins/PDFPlug.pm

@@ -27 +27 @@
 
 use ConvertToPlug;
+use unicode;
 
 sub BEGIN {
@@ -139 +140 @@
     my $outhandle=$self->{'outhandle'};
 
+    my $textref=$_[0];
+
     if ($self->{'use_sections'}
     && $self->{'converted_to'} eq "HTML") {
 
     print $outhandle "PDFPlug: Calculating sections...\n";
-    my $textref=$_[0];
 
     # we have "<a name=1></a>" etc for each page
@@ -196 +198 @@
     }
 
+    # turn any high bytes that aren't valid utf-8 into utf-8.
+    unicode::ensure_utf8($textref);
+
     print $outhandle "PDFPlug: passing $_[3] on to $self->{'converted_to'}Plug\n"
     if $self->{'verbosity'} > 1;
@@ -215 +220 @@
 
     # Add NumPages metadata (we have "<a name=1>" etc for each page)
-    my $textref = $_[0];
     my @pages = ($$textref =~ /\<a name=\d+\>/ig);
     $doc_obj->add_utf8_metadata($cursection, "NumPages", scalar(@pages));